Is global warming caused by human activity?

Paul Drallos, Ph.D. (Physics) July 9, 2008 (updated 10/24/09)

A great number of facts and declarations are often presented with regard to global warming and Mankinds role in it. Most of these assertions are derived from the following clear and simple line of reasoning: Fact: Fact: CO2 is a greenhouse gas whose presence in the atmosphere can cause Earths temperature to increase. Since the beginning of the industrial age, a. human activity has produced large amounts of CO2 b. atmospheric levels of CO2 have been increasing. c. global temperatures (on average) have been rising.

Based only on these facts, it is reasonable to conclude that post-industrial age human activity is the cause of recently observed global warming. It is also reasonable to speculate that continued production of CO2 , at current or higher rates, will continue to drive the temperature still higher, perhaps to catastrophic results. However, a conclusion based only on these facts is, in fact, a good example of how a little knowledge can be a dangerous thing. To understand this, lets take a moment to consider how Science works and what is required by it for a valid scientic conclusion.

How Science works

Initially, a scientist starts out with a hypothesis which is often formed on the basis of observations. However, the hypothesis may come from virtually anywhere - a dream, a guess,...it doesnt matter. The point is, once a hypothesis is formed, the rst task is to try to disprove it (falsify it.) If the scientist is unable to falsify the hypothesis, a next step is to publish it so that other scientists can attempt the falsication process. If their attempts at falsication also fail, then the hypothesis earns a measure of credibility. Furthermore, if the hypothesis represents a new or alternate explanation for a phenomenon which already has an established theory or explanation, there is an additional hurdle that

the new hypothesis must clear. The new hypothesis must show that the established theory is insucient to explain some observable phenomena which the new hypothesis can explain and the new hypothesis must also be able to explain or be consistent with the all of the other phenomena which the earlier theory could successfully address. Essentially, the bar is higher for new explanations. One thing a scientist does not do is try to prove that a hypothesis is correct. That would be the nature of pseudo-science and represents a logical error known as improper induction. This is because it is easy to obtain evidence in favor of virtually any hypothesis, but all it takes is one contradicting example to invalidate the conclusion. This concept of falsication is a central theme of Karl Poppers Theory of Demarcation and was also eloquently phrased by Albert Einstein when he said, A consensus of 100 scientists is undone by one fact.

The global warming hypothesis

Science also recognizes certain phenomena to be true, even if we dont fully understand the processes responsible for the observed phenomena. For example, you may know that magnets attract or repel each other without understanding the physics of magnetism. You may know that the Earth has a moon without knowing where the moon came from. Similarly, we know that the Earth has undergone many global temperature uctuations long before humans existed. Even though we dont know or understand the details responsible for those changes, we know they were the result of natural processes and not caused by Man. Hence, we know that the ability for natural processes to cause global temperature changes is beyond refute. Therefore, in order to justify the new proposition: That recent global warming is not the result of natural processes but is instead the result of human activity, it must be shown that 1. Global warming is taking place. 2. Natural causes cannot account for the observed warming. (established theory fails) 3. Human-activity can account for the observed warming (new theory succeeds,) plus, the new theory must be consistent with historical data. Each of these points must be true in order to conclude that human activity is the cause of recent global warming. Lets consider these three points, one-by-one...

1. Global warming is taking place true or false?

The answer to this question depends on the period of time we consider. For example, over the last 2000 years1,2 the temperature trend has been generally downward (cooling.) Over the last 700 years the trend has been constant and during the 19th and 20th Centuries, the trend has been upward. The current warming trend began about 1800. At that time, the Earth had just come out of a relatively cold period we call the Little Ice Age. According to the U.S. National Oceanographic and Atmospheric Administration (NOAA), National 2

Climatic Data Center3 , the temperature trend for the 20th Century (1895 to present) was 0.1 F per decade. This trend, however, was not a straight-line, steady trend. In fact, most of the 20th Century warming occurred between 1910 and 19404 . That was followed by a cooling trend from 1940 to 1975 (which resulted in world-wide fear of an impending Ice Age.) That cooling was then followed by another warming trend through 1998. However, during the last decade, 1998 to 2008 there has been a net cooling of 1.1 F (0.6C)5 . The recent cooling has brought temperatures back to their 1980 levels, which is where we were shortly after the 1940 - 1975 cooling. However, the large drop during 2007-2008 was probably anomalous and not a trend. Meanwhile, there has clearly not been any upward trend during the last decade while the trend has been slightly downward.

Fig. 1

Global temperature anomaly from 1988 to 2008.

So, is global warming taking place? Over the last ten years, there has been global cooling. The last hundred years has seen warming. The last 2000 years has been cooling, The last 10,000 years has been cooling. The last 200,000 years has been warming. Its a matter of choosing a time-scale. A 30-year trend was long-enough to prompt fears of a coming Ice Age (which turned out to be short-lived.) Then, after only a ten-year warming trend, new concerns about catastrophic warming began (and that warming only continued for another ten years, to 1998). Now we are ten years into a new cooling trend. Clearly, a few tens of years are not enough to establish a long-term trend. It is probably fair to say that the last 200 year warming trend is a recovery from the Little Ice Age and that the warming was a real trend. But we cannot say if more warming is in store or if the present cooling trend will continue. So it is impossible to say, at this time, if there is global warming or not. Therefore, Point 1 stands as neither true nor false. (Remember, Point 1 must be true in order to conclude that human activity is the cause of global warming.)

2. Natural causes cannot account for the present warming true or false?
For the purpose of this discussion, we will dene natural causes as any causes which are not the result of human activity. In the context of global warming, human causes are any 3

phenomena which result from large amounts of CO2 produced by human activity. This limits the inuence of human activity on global warming to within the last hundred years when quantities of human-produced CO2 became signicant. By looking at climate variations which occurred longer ago than a hundred years, we can gain insight about natural variations and natural causes. Then we can determine if the current climate behavior is within the domain of natural variation. On very long time scales6 , as shown in Fig. 2, we see that there have been at least four warm and cold periods of major proportions. The warm periods are characterized by temperatures about 10 C warmer than we see today. The warm periods are typically about 100 million years in length. Between the warm periods are the major ice ages, which typically last a few million years. The ice ages feature temperatures which are similar to what we see today. In fact, most geologists believe that we are still in the the midst of the most recent ice age. The major ice age cycles are believed to be related to changing land mass positions (Continental Drift) which aect the ocean currents and distribution of heat. From the gure, note that there is essentially no correlation between atmospheric CO2 levels and the major Ice Age cycles.

Fig. 2

The average global temperature and atmospheric CO2 for the last 600 million years6 .

On a somewhat shorter time scale, like millions or hundreds of thousands of years, we can consider the glacial and interglacial periods which have occurred within the present Ice Age. These periods correlate well with the Milankovitch Cycles which are related to variations in Earths orbit, rotation and tilt. Although the correlation of the Milankovitch Cycles with climate variation is very good, the variations in solar ux due to the orbital variations are not, by themselves, sucient to to account for the observed glacial/interglacial temperature variations. However, this has been explained through a positive feedback mechanism: Longer 4

snow seasons resulting from Milankovitch cooling bring about increased reectivity which add to the cooling and further increases the reectivity, etc., and is sucient to lead to the observed results. On somewhat shorter time scales (tens of thousands of years) we nally see some correlation between temperature and CO2 . But in all cases, we see that changes in CO2 follow changes in temperature by about 800 years. Because changes in temperature occur before changes in CO2 levels, changes in CO2 levels cannot possibly be the cause of the observed changes in temperature. However, it is simple to understand why temperature controls CO2 and why there is a several hundred year delay. It is explained by the fact that the solubility of CO2 in water is inversely proportional to temperature (see Fig.3). When the oceans warm, they release CO2 . When the oceans cool, they absorb CO2 . You can demonstrate this principle yourself by pouring a room-temperature glass of soda and leaving it sit at room-temperature, and also pouring a cold glass of soda and letting it sit in the refrigerator. The warmer glass will lose its zz within minutes while the colder glass retains much more. The zz is CO2 dissolved in the water. Furthermore, since the oceans have a very large heat capacity compared to the atmosphere, it takes hundreds of years for the ocean temperatures to respond to changes in atmospheric temperatures.

Fig. 3

The solubility of CO2 in water as a function of temperature.

On still shorter timescales, we can consider solar variability. There are several dierent solar cycles, typically on the order of tens of years. Like the Milankovitch cycles, there is a very good correlation with temperature and solar activity. Since the correlation is good and since it is not possible for the Earths temperature to aect solar activity, it must be true that solar activity aects Earths temperature. An aspect of this theory which is not completely understood is that variations in solar luminosity are not, by themselves, sucient to account for observed changes in temperature. Therefore, a feedback mechanism must also be involved. One such mechanism has been proposed which is based on the observation that increased 5

solar activity is also accompanied by increased solar wind which shields the Earth from cosmic rays. Cosmic rays, meanwhile, have been shown to play a role in cloud formation. So when the sun is more active, the solar wind deects more cosmic rays. This leads to a decrease in low-altitude cloud formation, permitting more sunshine to strike the Earth. This feedback appears to be sucient to account for the remaining temperature variation. However, it is irrelevant whether this is the correct feedback mechanism or not if the fact remains of a strong correlation between solar/sunspot activity and arctic temperature.

Fig. 4

The Arctic surface air temperature, solar irradiance and world hydrocarbon use.

The data in Fig. 4 show that solar irradiance correlates well with Arctic temperature while hydrocarbon use does not. Although this positive correlation between temperature and solar irradiance does not necessarily prove that solar irradiance is the cause and temperature is the eect, to argue otherwise, one must oer a physical mechanism by which temperatures in the Arctic are able to aect the behavior of the sun. Without such a mechanism, we must conclude that either the variations in solar activity are driving the observed temperature trends in the Arctic or that there has been a 120 year coincidence between Arctic temperatures and solar irradiance. So far, weve shown what we already knew from the beginning, that natural processes can cause global warming. However, we still have to answer the question as to whether natural causes can explain the currently observed warming trend. If the current temperature trend can be shown to be outside the limits of what we know to be natural variability, then a strong case can be made for an un-natural or man-made cause. So lets look at the recent trends compared to the past natural behavior.

Fig. 5 Climatic cycling over the last 16 000 years. The trends are indicative of both warming and cooling, depending upon the chosen starting point, and all except E are statistically signicant.

Within the context of historical temperature patterns, as shown in Fig. 5, there does not appear to be anything whatsoever unusual or beyond the limits of natural variability in the current temperature patterns. Therefore we must conclude that natural phenomena are capable of explaining the presently observed warming. So Point 2 is false. (Point 2 must be true in order to conclude that human activity is the cause of global warming.)

3. Human activity can account for the present warming true or false?
As weve noted earlier, there has been an approximately 0.1 F per decade rise in temperature over the last hundred years. Could this be the result of human activity? To address this question, let us rst appeal to some basic rules of logic so that we may quickly and easily disqualify some obviously invalid lines of reasoning.

disqualied evidence
Evidence of warming is often presented as evidence of human-caused warming, The logical construction of this argument is as follows: If human activity causes global warming, then the Earth will warm. The Earth is warming. Therefore, human activity causes global warming. This argument is invalid because the conclusion is drawn from a common logical error known as the converse error. As a clarifying example of a converse error, consider the argument: If Joe drops the glass, the glass will break. The glass is broken. Therefore Joe dropped the glass. 7

The conclusion that Joe dropped the glass is invalid because the glass might have been dropped by anyone else, or it could have been broken by some entirely dierent means. By the same logic, evidence of warming is simply evidence of warming. No conclusion about the role of human activity can be drawn from it. Another invalid argument is the claim that, If there is a consensus of scientists that global warming is human-caused, then human-caused global warming must be true. This argument is invalid for at least two reasons: First of all, the premise is not true. There is not a consensus of scientists as proved by a recent petition of 32,000 scientists who object to the premise. However, even if there were a consensus, it is still a logically invalid conclusion by reason of improper induction. Consider, for example, the consensus of scientists long ago who believed the Sun and stars revolve around the Earth. Clearly, consensus doe not make it true. Computer climate models do not provide evidence of human-caused global warming. There are many reasons to discount the predictions of the computer climate models. Most importantly, is the fact that the computer models are not real and they have not been validated against real observations. The validations have consisted of comparing one computer model to another. Meanwhile, there are signicant uncertainties at all levels of the models. These uncertainties include the primary energy source (the Sun,) physical processes (such as cloud formation and their role in climate, the values of many physical parameters, and the boundary conditions. In addition, limitations of computer resources prevent the use of small enough grid-size to resolve many important features. The system is extremely non-linear and sensitive to all of the above mentioned uncertainties and limitations. Any one uncertainty is sucient to produce unreliable results.

valid evidence
The fundamental mechanism through which human activity is believed to aect global temperature is based on the greenhouse gas eect of atmospheric CO2 . This eect is combined with the additional premise that human activity produces quantities of atmospheric CO2 that are, presumably, large enough to aect global temperature (through the mechanism of greenhouse gas warming.) This is a valid line of reasoning, and should be examined. Greenhouse gas theory is often explained with an overly simplied (and incorrect) model. The typical overly simplied model is represented in Fig. 6, from the British National Space Center. This model depicts the cooling of the Earth by thermal radiation from the surface. One of the reasons this model is incorrect is because the surface of Earth is not cooled primarily by thermal radiation.

Fig. 6 Overly simplied model of greenhouse gas phenomenon.

Fig. 7 More realistic greenhouse model. Lighter shading represents reduced opacity due to diminishing water vapor density.

A more accurate (but still simplied) greenhouse gas model is depicted in Fig.7. In this model, the eect of water vapor, the main greenhouse gas, is strongest near the Earths surface in the tropics. The eect decreases sharply with both altitude and latitude. The dark bands indicate the largest eects while the lighter bands indicate diminishing eects as latitude and altitude increase. In the tropical latitudes near the surface, there is sucient greenhouse opacity that the surface cannot eectively cool by thermal radiation emission. As a result, heat is conducted away from the surface by uid motions (shown by solid lines in Figure 7.) These motions carry the heat up in altitude and in latitude (toward the poles) to levels where it is possible for thermal radiation emitted from these levels to escape to space, shown by the wavy line. This is how the standard greenhouse gas model works. Indeed, all climate models predict that, if greenhouse gases are driving global warming, there will be a unique ngerprint of the eect in the form of a warming trend in the tropical troposphere which increases with altitude. Furthermore, since climate changes due to solar variability or other known natural factors will not show this characteristic pattern, this is a good test of the greenhouse hypothesis. Putting in some numbers, Lee et al. (2007) investigated four dierent General Circulation Models to examine how temperature, according to the models, should change with CO2 . Accordingly, and for all four models, the warming trend at high altitude in the tropics should be two to three times larger than the trend at the surface. We have good observational measurement of the temperature in this region of the troposphere. The records show that, at least since the 1960s, the temperature trend in the troposphere is about three-quarters of the surface trend. It is not 2 to 3 times larger, as predicted by the greenhouse models. From this, we must conclude that either the greenhouse model is fundamentally wrong, the surface temperature measurements are in error, the troposphere temperature measurements are in error or greenhouse warming is much smaller 9

than expected. If the physics is good and the measurements are good, we must conclude that the warming is not driven by the greenhouse eect, and therefore not related to human activity. If the greenhouse theory is true, global temperatures should rise in response to increases in atmospheric CO2 . However, looking back at Fig. 4, we see that carbon use was increasing exponentially during the period 1940 to 1970, yet the Arctic temperature was declining during that same period. Also, as described in 2, for the past several hundred thousand years, changes in atmospheric CO2 levels followed changes in temperature by a period of several hundred years. This behavior is not consistent with CO2 causing changes in temperature. Furthermore, if CO2 were the primary driver of temperature, then temperature would increase with increasing CO2 . But we also know that increasing temperatures would release even more CO2 from the oceans. That, in turn, would drive the temperatures higher still. By the nature of this logic, temperatures would never come down. This situation is contrary to the historical record which features many cycles in which temperatures and CO2 levels have risen, and then then fallen. In this section, we have shown that 1.) The expected physical evidence for greenhouse gas warming is not present; 2.) The expected cause and eect relationship between temperature and CO2 is not present; 3.) The greenhouse model is inconsistent with established historical cyclical patterns. Based on any of these three observations, we must conclude that the statement Human activity can account for the present warming, is not true.

Conclusion
Recall our conditions that, in order to justify the new proposition that recent global warming is not the result of natural causes but is, instead, the result of human activity, the following conditions must be true: 1. 2. 3. Global warming is taking place Natural causes cannot account for the observed warming. Human activity can account for the observed warming and the new theory must be consistent with historical data

What weve shown in the previous sections is that Condition 1 is not true. Condition 2 is not true and Condition 3 is not true. Since all three of these conditions must be true in order to justify the theory of human-cause global warming, we must conclude that there is no logical basis on which to believe that theory. One can also look at this situation from a more practical point of view. On the one hand, we have the natural explanation. There is good correlation between temperature and Earths orbital variations. There is good correlation between temperature and solar variations. There is historical evidence of natural climate cycles. There is historical evidence that changes in temperature cause changes in atmospheric CO2 . There is scientic basis for each of these observations. On the other hand, we have the anthropogenic greenhouse gas explanation.

10

The physical evidence for greenhouse gas warming is absent. The correlation between CO2 and temperature is absent. The historical evidence is not consistent with CO2 acting as a driver of temperature. Temperature trends during the most recent decade have displayed the exact opposite behavior than predicted by the greenhouse theory. In light of these observations, by what logic can we justify choosing a new theory which has no supporting physical evidence, no positive correlation with observations, inconsistency with historical evidence and currently predicts the opposite of what we now observe? Furthermore, by what logic do we choose such a theory over an established theory that has physical evidence and positive correlation with observations, is consistent with the historical record and currently predicts what we now observe? (We are currently witnessing a decrease in solar activity and a corresponding decrease in global temperature, as expected.) Only irrational logic could support such a choice.

References
1. Davis, J C and Bohling, G C, 2001. The search for patterns in ice-core temperature curves, in Geological Perspectives of Global Climate Change (ed: L C Gerhard et al), American Association of Petroleum Geologists, Studies in Geology, 47:213-229 2. NSIDC User Services, 1997. The Greenland Summit Ice Cores, CD-ROM, GISP-2/GRIP, World Data Center A for Glaciology, CIRES, University of Colorado, Boulder. 3. Climate Summary April 2008, NOAA Satellite and Information Service, National Environmental Satellite, Data and Information Service, http://www.ncdc.noaa.gov/oa/climate/research/cag3/na.html 4. National Climatic Data Center, Global Surface Temperature Anomalies (2007) http://www.ncdc.noaa.gov/oa/climate/research/anomalies/anomalies.html and NASA GISS http://data.giss.nasa.gov/gistemp/graphs/Fig.D.txt. 5. Met Oce Hadly Center http://hadobs.metoce.com and Climatic Research Unit, University of East Anglia, http://www.cru.uea.ac.uk/cru/data/temperature/ 6. Monte Hieb, Climate and the Carboniferous Period: Similarities with Our Present World, Plant Fossils of West Virginia (Web site), updated September 19, 2006; available at http://mysite.verizon.net/mhieb/WVFossils/Carboniferous climate.html. For temperature data, see C.R. Scotese, Climate History: Ice House or Hot House? PALEOMAP Project, April 20, 2002; available at http://www.scotese.com/climate.htm. For CO2 data, see Robert A. Berner and Zavareth Kothavala, Geocarb III: A Revised Model of Atmospheric CO2 over Phanerozoic Time, American Journal of Science, Vol. 301, February 2001, pages 182-204; available at http://www.ajsonline.org/cgi/content/abstract/301/2/182.

11